A corpus-centered approach to spoken language translation

Abstract

This paper reports the latest performance of components and features of a project named Corpus-Centered Computation (C'3), which targets a translation technology suitable for spoken language translation. C3 places corpora at the center of the technology. Translation knowledge is extracted from corpora by both EBMT and SMT methods, translation quality is gauged by referring to corpora, the best translation among multiple-engine outputs is selected based on corpora and the corpora themselves are paraphrased or filtered by automated processes.

Design, development and field evaluation of a Spanish into sign language translation system

“…itl.nist.gov/iad/mig/tests/mt/). The best performing translation systems are based on several types of statistical approaches [3,21,28,31], including example-based methods [9,36], finite-state transducers [7] and other data-driven approaches. The progress achieved over the last years has been thanks to several factors such as efficient algorithms for training [29,39], context-dependent models [40], efficient algorithms for generation [19,39], incorporation of more powerful computers and bigger parallel corpora, and automatic error measurements [1,2,30].…”

4-Camera model for sign language recognition using elliptical fourier descriptors and ANN

“…Gesture recognition systems that are implemented by statistical approaches [19], example based approaches [20], finite state transducers [21] have registered higher translation rate.…”

Automatic categorization for improving Spanish into Spanish Sign Language machine translation

“…Vermobil, Eutrans, LC-Star, PF-Star and, finally, TC-STAR and EuroMatrixPlus) and in the USA (GALE). As regards the evaluation campaigns organized by NIST (National Institute of Standards and Technology) in the USA and the EuroMatrixPlux FP7 project in EU, the best performing translation systems are based on various types of statistical approaches Mariño et al, 2006), including example-based methods (Sumita et al, 2003), finitestate transducers (Casacuberta and Vidal, 2004) and other data driven approaches. The progress achieved over the last 10 years is down to several factors such as efficient algorithms for training (Och and Ney, 2003), context dependent models (Zens et al, 2002), efficient algorithms for generation (Koehn, 2003), more powerful computers and bigger parallel corpora, and automatic error measurements (Papineni et al, 2002;Banerjee and Lavie, 2005;Agarwal and Lavie, 2008).…”